Abstract
Noncoherent demodulation is an attractive choice for many wireless communication systems. It requires minimal protocol overhead for carrier synchronization, and it is robust to radio impairments commonly found in low-cost transceivers. Machine learning techniques, such as neural networks and deep learning, offer additional benefits for these systems. Practical communication systems often include nonlinearities, non-stationarity, and non-Gaussian noise, which complicate mathematical derivation of optimum demodulators. Learning approaches can optimize demodulator performance directly from simulated or measured radio data, which is often plentiful in the design and verification of today’s integrated transceivers. This paper examines several candidate neural network topologies for use in noncoherent demodulation and provides a mathematical framework for their comparison. Each is based on a complex-valued feature detection layer, which may be characterized as coherent or noncoherent, followed by one or more real-valued classification layers. Backpropagation equations for the noncoherent feature layer include a synchronization term that facilitates training with noncoherent input data. The coherent layer does not synchronize training data, however a noncoherent demodulator can still be constructed by increasing the coherent layer capacity and adding a max pooling layer to marginalize the unknown signal phase. A frequency classification example highlights the differences between the topologies and confirms that optimum noncoherent demodulation can be learned in the presence of AWGN and random phase offsets. The topologies considered here are suitable for noncoherent demodulation of power-efficient modulations such as FSK and ASK, which are typical in today’s short-range wireless communication systems. It is hoped that such topologies will lead to a future common architecture that can support the wide range of modulation formats in this space.
Highlights
N EURAL networks represent a flexible, adaptable architecture for signal classification tasks such as demodulation and detection
This paper considers the selection of neural network topologies suited to noncoherent demodulation of power-efficient modulations such as frequency shift keying (FSK), Gaussian-filtered FSK (GFSK), minimum shift keying (MSK), amplitude shift keying (ASK), and orthogonal M-ary modulation (M-orth)
These results show that the noncoherent feature detection layer defined in (1), when combined with a real-valued classification layer, produces an efficient classifier well-suited to noncoherent demodulation and capable of learning optimum performance
Summary
N EURAL networks represent a flexible, adaptable architecture for signal classification tasks such as demodulation and detection. Communication signals are often exposed to nonlinearities, time variance, non-Gaussian noise, and other impairments that make mathematical derivation of optimal classifiers intractable Recasting these problems in terms of machine learning allows demodulators to be optimized directly from data. The growth in computing power in recent decades has led to very detailed and accurate modeling capability for communication channels and radio impairments, and actual system performance is typically in close agreement with simulations. This has become an important tool in the design of today’s wireless standards and ASIC transceiver hardware.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
